Data Mining Assignment #1
CSC592 – Fall ‘05
Overview
You are to construct a decision tree for the given data set (tennis.arff – see course
website) using the ID3 decision tree algorithm in the Weka toolset. The dependent
variable for the tree is ‘play’. Once you have constructed the tree use the tree and the
data Weka provides about the tree to answer the questions below.
Questions
1. Does the tree adequately describe the data? Why? Why not?
2. Use the decision tree to figure out the value for the dependent variable for the
following instance:
outlook
rainy
temperature
hot
humidity
high
windy
false
play
?
Justify your answer!
Instructions






Start the Weka explorer.
Load the tennis.arff file using the ‘preprocess’ tab.
Switch tabs to ‘classify’ and select the ID3 algorithm with the ‘choose’ button
(you will find ID3 under trees).
Set the ‘test options’ to ‘use training set’.
Make sure that ‘play’ is the attribute that shows below the test options.
Hit the ‘start’ button – the big pane should display information about the decision
tree built.
Grading
Each question is worth 5 points.
Handing in your assignment
Hand in a written report describing your results. The due date is September 21st in class.